Co‐Electroporation with Xenopus Laevis Oocytes Reprograms Normal and Cancerous Human Cells to Resemble Reprogramming Normal and Cancerous Human Cells to Resemble induced Human Pluripotent Stem Cells

The co‐electroporated human cells formed colonies on irradiated mouse embryonic fibroblast cells (all study cells) and StemAdhereTM substrate (assessed for co‐electroporated buccal mucosa cells). Cells derived from BMSC, BJ cells, HPA, CD4TLs, and buccal mucosa cells expressed Oct 3/4, Nanog, SOX‐2, Rex‐1, TRA‐1‐60, and SSEA‐1. Cells derived from co‐electroporated HeLa cells expressed Oct 3/4; cells derived from co‐electroporated MCF‐7 cells expressed Oct 3/4 and Nanog. Reprogramming efficacy for CD4TLs was 23.4 ± 3.5%. Co‐electroporated HPA trans‐differentiated into neural progenitor cells in culture conditions that foster neural differentiation. Control experiments suggested that the electroporate conveyed a reprogramming factor(s).

Many species have evolved mechanisms for cellular reprogramming; pathways for inducing cellular dedifferentiation and redifferentiation exist a wide variety of organisms including some species of bacteria, [19] plants, [20,21] and lower animals [22]. The poor efficacy of reprogramming achieved using current approaches may be due the use of methods that lack vital reprogramming components naturally present in some living organisms.
In the present study, Xenopus laevis oocytes were chosen as a source of natural reprogramming factor(s) based on the successful reprogramming events reported for mammalian somatic cells induced by Xenopus laevis egg extracts [23]. We show that co-electroporation of living Xenopus laevis oocytes with normal human cells and cancerous human cells lines reprogrammed these cells to resemble iPSc with respect to colony morphology and expression of stem-cell associated transcription factors recognized as markers of pluripotency. Furthermore, this co-electroporation technique achieved a relatively high level of reprogramming efficiency.

Materials and Methods
The human subjects who provided buccal mucosa tissue samples (using a non-invasive technique) gave written informed consent. Procedures involving Xenopus laevis were conducted in accordance with published laboratory standards [24].

Preparation of Xenopus Laevis Oocytes
Female, egg-bearing Xenopus laevis (NASCO) were at kept at ˜18°C using a 12 / 12-hour light/dark cycle in carbon-filtered water supplemented with 13.3 g/gallon NaCl, which was changed daily [24].
Prior to surgical removal of oocytes, frogs were anesthetized in a plastic beaker containing 1 L of 0.2% tricane solution (Sigma) for up to 20 min and, then, placed on a dissecting pan filled with ice. After a 0.5 cm incision through the skin and muscle layers, the bags of ovaries were removed and placed into oocyte washing (OW) solution (82.5 mM NaCl, 5.0 mM 2-[4-(2-hydroxyethyl) piperazin-1-yl]ethanesulfonic acid [HEPES], 2.5 mM KCl , 1 mM MgCl 2 , 1.0 mM Na 2 HPO 4 , and 0.5% penicillin/streptomycin [pen/ strep] at pH 7.4 [penicillin and streptomycin from Gibco; others from Sigma]). Bags containing ovaries were opened with fine forceps, the ovaries were rinsed several times in OW, and treated with a 0.2% collagenase type II solution (Worthington Biochemical Corporation) for ≥ 1 hour at room temperature. The defolliculated oocytes were rinsed in OW solution and incubated overnight in fresh holding buffer (HB) containing 5 mM NaCl, 5.0 mM HEPES, 2.5 mM KCl, 1 mM MgCl 2 , 1.0 mM Na 2 HPO 4 , 0.5% pen/strep, 1.0 mM CaCl 2 (Sigma), 2.5 mM sodium pyruvate, and 5% heat-inactivated horse serum (Sigma) titrated to pH 7.4. Recovered oocytes in the final stage of maturity were collected in sterile 6-well cell culture clusters (Costar) prefilled with an HB solution and incubated at 17°C for 24 hours before electroporation experiments.

Cell Lines
Human bone marrow stromal Cells (BMSCs) and stably transfected GFP-expressing BMSCs (BMSC GFP ) were provided by Tulane University Center of Gene Therapy. Prior to release from the source, two trials of frozen, passage-1 cells were analyzed over three passages for colony forming units, cell growth, and differentiation into fat, bone, and chondrocytes. The BMSC and BMSC GFP were cultured in Dulbecco's modified Eagle's Medium (DMEM; Sigma), supplemented with 10% fetal bovine serum (FBS; Gibco) and 1% streptomycin/penicillin (Gibco) and cultured in 25 cm 2 flasks at 37°C with 5% CO 2 . At day 4, the cultures were washed with phosphate-buffered saline (PBS; Sigma) to remove the non-adherent cells and further expanded until ~8 0% confluence, when they were harvested and expanded in 75 cm 2 flasks.
Human buccal mucosa cells were obtained from healthy human subjects approximately 1 hour before the co-electroporation procedure. Subjects abstained from drinking coffee for 1 hour before collection. Subjects' mouths were rinsed twice with Listerine® followed by sterile distilled water before swabbing. Cells were collected by swabbing firmly on the inside of the cheek 20 times on both sides using a MasterAmp™ Buccal Swab Brush (Epicentre Biotechnologies). The brush holding cheek cells was placed into a 50 mL centrifuge tube filled with 20 mL of sterile filtered PBS (Sigma) containing 1% pen/strep. The sample was vigorously twirled for 30 sec and, then centrifuged at 200g for 7 min. Pelleted cells were resuspended in 5 mL of serum-free DMEM (ATCC) supplemented with 1 mM sodium pyruvate, 0.1 mM NEAA, and 1% pen/strep. Buccal mucosa cells were kept in a refrigerator at 4⁰C before use.

Co-Electroporation of Xenopus Laevis Oocytes with Human Cells
Electroporation parameters for Xenopus Laevis oocytes were developed from several published studies of electroporation [25][26][27]. Forty to fifty fresh oocytes from suspensions with ≥ 90% viability (oocytes showing abnormal pigment distribution or signs of damage of equatorial band, patchy gray membranes during the defolliculation process were discarded) were placed in sterile Gene Pulser electroporation cuvettes (Bio-Rad) prefilled with 400 μL of serum-free DMEM containing 1.0 x10 5 -1.5x10 5 cells/mL of human cells in suspension. Cuvettes were filled to 800 μL with serum-free DMEM and then placed into the shocking chamber. Co-electroporation of frog oocytes with the suspension of human cells was conducted using the following parameters: 150 v/cm / 25 μF / 7 pulses, with time constant at 0.5 -0.7 msec. After electroporation, cuvettes containing oocytes and the human cells were incubated at 17 o C for three hours to recover. The human cells were transferred to T25 culture flasks containing iMEF feeder cells for culturing.

Culturing of Human Cells Following Co-Electroporation
The co-electroporated human cells were cultured at 37 o C on iMEF feeder cells in 0.1% gelatin-coated (gelatin from Sigma) T25 culture flasks containing 5 mL of specially formulated Embryomax® DMEM culture medium (Millipore). Medium was supplemented with 15% FBS, 1 mM sodium pyruvate, 0.1 mM NEAA, 1% pen/strep, 100 μM beta-mercaptoethanol (Gibco), and 1000 U/mL ESGRO® (Millipore). To maintain the cells in an embryonic stem cell-like state, 1000 U ESGRO® per 1.0 mL of tissue culture media was required. After formation of clusters, the human cells were separated from the feeder cells using the differential sedimentation technique previously described by Doetschman,[28] which removed > 99% of contaminating feeder cells from the electroporated human cell suspension. Trypsinized (trypsin from Sigma) human cell cultures containing iMEFs were centrifuged at 200 g, resuspended in 10 mL of complete ES culture medium, and transferred to a new T25 cell culture flask for 30 minutes at 37°C. Following incubation, the culture medium containing mostly human cells was transferred to a new T25 culture flask for 1-hour at 37°C to remove all remaining fibroblast feeders. Following the second incubation, the culture medium containing the human cells was removed, and the cells were counted, centrifuged again at 200 g, and resuspended in the ES culture medium.

Subculturing
After separation from the feeder cells, the human cells were placed on T25 culture flasks containing either iMEF feeder cells or feeder-free StemAdhere™ pluripotency substrate (Primorigen Biosciences). Subcultured human cells were grown in NutriStem™ (StemGent).

Calculation of Reprogramming Efficacy
Fluorescent immunohistochemically detectable expression of the Nanog gene by cells derived from CD4TLs occurred between 12 h -24 h following co-electroporation with Xenopus laevis oocytes. This expression preceded the formation of tight iPSc-like clusters, making it possible to determine the efficiency of reprogramming by calculating the proportion of cells expressing Nanog gene. The mean for the reprogramming efficiency was calculated by counting the total number of Nanog-positive cells per specimen in each T25 flask (3-4 times), subtracting the number of nonspecific binding sites in the control flasks, dividing by the original number of cells having undergone co-electroporation and multiplying by 100%. The standard deviation of the mean was also calculated.

Cryopreservation of Reprogrammed Cells
Cells were cryopreserved using a standard slow-cooling freezing method [29]. One mL of cells was gently resuspended in 1.5 mL cryovials (Nalgene) containing 0.5 mL of 2X hES cell freezing medium (60% FBS, 20% hES cell culture medium, and 20% dimethyl sulfoxide. Cryovials were transferred to 5100 Cryo 1°C Freezing Container (Nalgene), refrigerated at -80°C overnight and then rapidly transferred to liquid nitrogen refrigeration units.

Trans-Differentiation into Neuronal Progenitor Cells
After formation of clusters, reprogrammed cells derived from HPA were separated from the feeder layer using the Doetschman differential sedimentation technique,[28] and were dissociated enzymatically using collagenase IV (Sigma; 200 U/mL) for 30 min at 37 o C generating a cell suspension containing small cell aggregates and single cells. Cell culture conditions for growing neural progenitor cells (NPs) from embryonic stem cells were employed [30]. The cells were washed in warm Neurobasal A medium (GibcoBRL/Invitrogen), pelleted and resuspended in pre-warmed (37°C) standard human embryonic stem cell culturing medium (hESC) supplemented with following growth factors and neuronal and other supplements: fibroblast growth factor-2 (10 ng/mL), epidermal growth factor (20 ng/mL), 1% B27, 1% N2, 1% pen/strep, 1% l-glutamine, 1% non-essential amino acids (NEAA), 0.2% beta-mercaptoethanol, and 20% Knockout Serum Replacement (all media components from Gibco-BRL/ Invitrogen). The HPA-derived cells in suspension were then seeded at high cell density (150-200×103 cells/cm 2 ) onto BD BioCoat™ and laminin-coated 150mm petri dishes (Beckson Dickenson), and the medium was supplemented with hESC medium and 4ng/ mL fibroblast growth factor-2. Proliferating HPA-derived neural progenitors were observed in 8-10 days. The neural rosettes were dissociated by short (5-10 min) collagenase IV treatment into single cells and re-seeded under the same conditions, thus generating a monolayer population of proliferating neural progenitors.

Qualitative Assessment of Colony Morphology
Assessment of colony morphology (resemblance to iPSc colonies) was performed by Dr. Nikolai Strelchenko, PhD of the hESC Research Lab at Reproductive Genetics Institute, Chicago, IL, USA and Dr. Arshak Alexanian, VMD, PhD, of the Department of Neurosurgery, Neuroscience Research Laboratories, Zablocki Veterans Affairs Medical Center and of Medical College of Wisconsin, Milwaukee, WI, USA.

Alkaline Phosphatase Staining and Fluorescent Immunocytochemistry
Histochemical staining for alkaline phosphatase (AP) was conducted using the Vector® Blue Alkaline Phosphatase Substrate Kit III (Vector Laboratories, Inc.). Expression of several pluripotency factors was assayed using fluorescent immunohistochemistry conducted at room temperature. Samples from all populations of human cells in T25 culture flasks went through the following steps: (a) the growth medium was removed, (b) washed three times with PBS, c) fixed in -10⁰C methanol, c) washed three times with PBS, d) incubated for 20 min in 10 % normal serum, e) incubated for 60 min. in primary antibody diluted in 1.5% normal serum, f) washed three times with PBS, g) incubated for 45 min. in the dark with secondary antibody diluted in 1.5% normal serum, h) washed three times with PBS and left in 3 rd rinse, I) examined under an inverted-phase contrast fluorescent microscope, j) PBS replaced with the anti-fading reagent 2% DABCO (Sigma), and k) processed T25 flasks with specimens were sealed with parafilm, wrapped in aluminum foil and stored at 4⁰C.

Control Experiments
The controls described in (Table 1) were used to test for the effect of the presence of human cells, oocytes, feeder cells, co-electroporation, and the electroporate on reprogramming (expression of Nanog; detected using fluorescent immunohistochemistry). Table 1 lists findings from the control experiments conducted on all human cell types used in this study. Nanog was not detected in human cells from controls "a", "b", "c", and "f " A small number of human cells from control "d, " in which non-electroporated human cells were exposed for 3 hours to electroporate, expressed the Nanog gene (reprogramming efficiency of ˜0.4%; calculated only for CD4TLs). A similarly low number of human cells from control "e" expressed the Nanog gene (0.9% efficiency, calculated only for CD4TLs); in this control, human cells were electroporated in the absence of oocytes and then were exposed to electroporate for 3 hours.

BMSC and BMSC GFP
Within one week of co-electroporation with Xenopus laevis oocytes, cells derived from human BMCS strongly co-cultured with iMEF cells expressed the pluripotency-associated transcription factors Oct3/4, SOX-2, Nanog, Rex-1, and SSEA-1 and formed colonies resembling those known to form by iPSc in culture in culture (Figure 1).
*Approximately 10 5 of the following: bone marrow stromal cells, BJ cells, human pre-adiposites, CD4TLs, human buccal mucosa cells, HeLa cells, MCF-7 cells (control experiments conducted separately with each human cell type) ; †oocytes removed from the electroporate prior to incubation; ‡calculated using CD4TLs.
CD4TLs, human CD4+ T-lymphocytes; ES, embryonic cell; iMEF, irradiated mouse embryonic fibroblasts. In separate studies, BMSC GFP were co-electroporated with Xenopus oocytes and grown on iMEF cells. The resultant cell colonies resembled those of iPSc and contained cells emitting green fluorescence (data not shown).

BJ Cells
Co-electroporation in the presence of Xenopus oocytes, followed by co-culture on iMEF feeder cells, resulted in reprogramming of BJ cells, evidenced by a high level of alkaline phosphatase activity and resemblance to iPSc in colony morphology and the expression of Oct3/4, Nanog, SOX-2, TRA-1-60, Rex-1, and SSEA-1 (Figure 2).
One month after cryopreservation of the reprogrammed HPA-derived cells, the reprogrammed cells were thawed, resulting in 78% viability. By day 4 after subculturing on fresh feeder cells the reprogrammed HPA-derived cells formed secondary clusters resembling those formed by iPSc (data not shown).
Sub culturing cells derived from HPA following co-electroporation in conditions that promote the neural differentiation of embryonic stem cells resulted in formation of cells expressing various immature and mature neural markers including nestin, NCAM, B3T, and GFAP ( Figure 4).

CD4TLs -Reprogramming and Efficiency
Within 3 to 5 days after transfer to feeder cell layers following co-electroporation with Xenopus laevis oocytes, the human CD4TLs formed colonies similar to those formed by iPSc. Cells in these colonies had high levels of alkaline phosphatase activity ( Figure 5) and strongly expressed Oct3/4, Nanog, SOX-2, TRA-1-60, Rex-1, and SSEA- (Figure 6).       oocytes, the cells derived from human CD4TLs co-cultured with iMEF started to express the Nanog gene. During this time period, single cells and small iPSc-like clusters in which individual cells could be counted were present (data not shown). The proportion of cells expressing Nanog and the total number of cells were counted for calculation of reprogramming efficacy, which was 23.4 ±3.5%.

Human Buccal Mucosa Cells
Freshly obtained human buccal mucosa cells, co-electroporated in the presence of Xenopus oocytes and cultured on iMEF and on feeder cell-free StemAdhere™ substrate, gave rise to cells that formed colonies similar to those of iPSc (Figure 7). Cells in these colonies expressed Oct3/4, Nanog, SOX-2, TRA-1-60, Rex-1, and SSEA-1 (Figure 8).

HeLa and MCF-7 Cells
Two human cancer cell lines, HeLa and MCF-7, were subjected to co-electroporation with Xenopus laevis oocytes followed by co-culture on iMEF. The cells derived from co-electroporation of these tumor cells showed partial dedifferentiation, with formation of clusters and expression of Oct 3/4 (HeLa-derived cells and MCF-7-derived cells. Figure 9) and Nanog (MCF-7-derived cells; Figure 9). The cell clusters tended to be smaller than those derived from co-electroporation of non-tumor cells (data not shown).

Discussion
The limited ability of many human tissues to regenerate has spurred interest in methods to produce iPSc for therapeutic applications. We evaluated a new methodology for the non-viral reprogramming of cells into iPSc. Using co-electroporation of living Xenopus laevis oocytes with various human normal and cancer cell lines, we obtained cells resembling iPSc as evidenced by colony morphology and expression of human iPSc markers.
Human bone marrow stromal cells showed signs of reprogramming into cells resembling iPSc, with colony formation and strong expression the pluripotency-associated transcription factors Oct3/4, Nanog, SOX-2, TRA-1-60, Rex-1, and SSEA-1. The isolation of human BMSCs represents a routine procedure at many hospitals, and this new method for the generation of human BMSC-derived iPSc may present opportunities for their therapeutic applications in humans.
Because of the pioneering studies on successful retroviral vector-mediated reprogramming of fibroblast cell lines, [31] we tested reprogramming of BJ cells using Xenopus laevis oocyte co-electroporation. Cultured BJ cells were reprogrammed into cells that resembled iPSc. These cells formed clusters with high AP activity and strong expression of major stem-cell markers, namely Oct3/4, Nanog, SOX-2, TRA-1-60, Rex-1, SSEA-1. The successful reprogramming of BJ cells provided further evidence of the effectiveness of this reprogramming method.
We also tested this reprogramming method using HPA because these cells can be safely isolated. The HPA were purchased from a commercial source that obtained HPA from less than 50 g of human adipose tissue obtained from a clinic that performs liposuction. Co-electroporation with Xenopus laevis oocytes resulted in reprogramming the HPA into cells resembling iPSc. We also showed that these HPA-derived deprogrammed cells could be cryopreserved, thawed, subcultured, and trans-differentiated into cells expressing neural and neural progenitor markers. These data suggest that this reprogramming technology may have the potential for large-scale production of inexpensive human iPSc from adipose tissue.
Reprogramming of T cells to produce stem cells for adoptive transfer constitutes an important area of interest in immune-based oncology therapy [32]. While human CD4TL can be easily isolated from as little as 5 mL of peripheral blood, this specific cell type has not been a common research target for producing iPSc [33]. We demonstrated that our protocol reprogrammed CD4TLs into cells resembling iPSc with formation of clusters on iMEF feeder cells, high alkaline phosphatase activity, and expression of pluripotency-associated transcription factors.
The oral mucosa contains cells that can be obtained without invasive techniques, but the literature on reprogramming of these cells is limited. Miyoshi et al reported production of iPSc from oral mucosa cells (obtained from oral biopsy tissue) using retroviral transfer of Oct-4, Sox-2, c-Myc, and KLF4 [34]. Using co-electroporation with Xenopus laevis oocytes, we were able to reprogram cells from the buccal mucosa (obtained non-invasively) to resemble iPSc. Furthermore, these reprogrammed cells could be subcultured and were able to grow on the feeder-free StemAdhere™ substrate. The success with reprogramming of buccal mucosa cells suggests the possibility of generation of large amounts of human autologous stem cells from this easily obtained tissue.
The prospect for converting cancer cells into normal or benign quiescent cells using a reprogramming approach, which can alter cellular transcription programs, is widely discussed in the scientific literature. and viral-mediated transfer of stem cell transcription factors to reprogram colorectal cells [39]. Using co-electroporation with Xenopus laevis oocytes, we observed that cells from these human cervical carcinoma and breast adenocarcinoma cell lines partially de-differentiated. The cells formed iPSc-like clusters, with some cells expressing Oct 3/4. This partially reprogramming may provide a transitional point for potential redifferentiation into normalized cells. Partially reprogrammed cells may also be amenable to trans-differentiational reprogramming [40].
The relatively high rate of reprogramming achieved using his co-electroporation method, 23.4%, is of note. The efficiency of reprogramming reported in the literature includes 0.5% with standard, four-factor retroviral methodology, [41,42] 0.98 %-2.34% when adding two more reprogramming factors , [10] 2-4% with the use of dox-inducible lentiviruses , [12] and 18% with cell-to-cell extracts.
[43] Because our main objective was to test reprogramming using co-electroporation with Xenopus laevis oocytes in multiple types of human cell and tissues lines, we limited the evaluation of reprogramming efficacy to human CD4TLs, which could be evaluated for Nanog expression at early stages before formation of tightly clustered colonies. Modifications of the method may improve the reprogramming efficiency. For example, preliminary studies suggest that the reprogramming efficiency can be modulated by fluctuations in barometric pressure and environmental temperature (Bioquark Inc., data on file).
The absence of natural reprogramming signals may result in failure or inconsistency of cellular reprogramming. Compromise of signaling factors may occur during the preparation of crude extracts from Xenopus laevis oocytes, in which potentially vital nuclear and cytoplasmic components present in living eggs could be disrupted. Ganier et al observed similarly low efficacy of nuclear transfer and reprogramming of mouse embryonic fibroblasts using pretreatment with Xenopus laevis oocyte extracts and with viral-mediated expression of Oct-4, Sox-2, Klf4, and cMyc (OSKM). However, reprogramming efficiency was improved approximately 10-fold when extract pretreatment and viral transfer of the transcription factors were both performed [44]. Using a process that promotes the natural order of reprogramming signals also appears important. Grad et al reported that reprogramming that deviates from what is known of the normal sequence of events (induction of Nanog before OSKM) produced abnormal cells [45].
The results of control experiments strongly suggest that during co-electroporation of living Xenopus laevis oocytes, there is transfer of a vital, soluble reprogramming factor or factors to the human cells. This interpretation is supported by low but detectable level of reprogrammimg in controls "d" and "e", in which human cells that were not co-electroporated with Xenopus laevis oocytes were exposed to electroporate from Xenopus laevis oocytes. The possibility of multiple factors being necessary is logical considering such a biologically significant cellular event as reprogramming would be regulated through multiple factors and pathways. In addition, successful retroviral and other molecular reprogramming techniques require transfer of multiple transcription factors [31]. Identification, purification, and amplification of active reprogramming components transferred during co-electroporation may provide opportunities for investigation of therapeutic potential.
The results presented above need to be confirmed by independent studies, and further research is needed for full proof of concept of reprogramming cells to pluripotency. Ongoing activities include assessment of redifferentiation and trans-differentiation, molecular karyotyping, DNA fingerprinting, and teratoma formation. Biochemical and molecular analysis of the intrinsic molecular mechanisms underlying the Xenopus oocyte-mediated reprogramming phenomena are in progress.

Conclusion
The data obtained from analysis of human cells subjected to co-electroporation with Xenopus laevis oocytes show that this reprogramming system induced human somatic cell reprogramming in a fast, efficient, highly reproducible, standardized fashion in different cell types. The system also provides for easy separation of reprogrammed cells from oocytes and subculture of these cells in the absence of feeder cells.